Home › Blog › Tutorials
How to Convert Units Accurately (and the Mistakes That Cost You)
Unit conversion looks like primary-school arithmetic, and most of the time it is — right up until it is not. A spacecraft lost because one team worked in pounds of thrust and another in newtons. An airliner gliding to an emergency landing because its fuel was calculated in pounds and delivered in kilograms. Unit errors are rarely about hard maths; they are about habits. This guide sets out the method that professionals use, the special cases that break the naive method, and how to check your own conversions — with the converter pages on this site doing the arithmetic while you keep hold of the thinking.
The one method that always works
Every conversion between units of the same quantity is a multiplication by a factor that equals 1. Converting 12 inches to centimetres: 12 in × (2.54 cm / 1 in) = 30.48 cm. The inches cancel, the centimetres remain. This sounds pedantic until you chain it: convert 60 miles per hour to metres per second by multiplying by (1.609 km / 1 mile) and then (1000 m / 1 km) and dividing by 3600 seconds — 26.8 m/s, with every unit cancelling cleanly on paper. If the units do not cancel to the unit you want, the setup is wrong, and no amount of careful arithmetic will save it. That cancellation check is the single most valuable habit in this entire guide.
Two practical corollaries. First, convert in one step with a precise factor rather than rounding at every hop: inches → centimetres → metres with rounded intermediates drifts; inches → metres directly (× 0.0254, an exact definition) does not. Second, keep guard digits: carry at least two more decimal places through the working than you need in the answer, and round once, at the end.
Reference factors worth knowing
You do not need to memorise many. These cover most daily and technical work, and each links to a converter that shows the formula and a full chart:
- Length: 1 inch = 2.54 cm (exact); 1 foot = 0.3048 m (exact); 1 mile = 1.609344 km (exact). See metres to feet and miles to km.
- Mass: 1 pound = 0.45359237 kg (exact); 1 ounce = 28.3495 g. See kg to lbs.
- Volume: 1 US gallon = 3.78541 litres; 1 US fluid ounce = 29.5735 ml. Note the "US" — an imperial gallon is 4.54609 litres, 20% larger, and recipes and fuel figures cross the Atlantic badly.
- Speed: 1 mph = 1.609344 km/h; 1 knot = 1.852 km/h (exact).
- Pressure: 1 psi = 6.89476 kPa; 1 bar = 100 kPa exactly; standard atmosphere = 101.325 kPa.
The full library is on the conversions index, organised by quantity.
Temperature: the famous exception
Temperature scales do not share a zero, so temperature is not a ratio conversion and the multiply-by-a-factor method fails. Celsius to Fahrenheit is F = C × 9/5 + 32: the × 9/5 resizes the degree, the + 32 shifts the zero. The classic trap is temperature differences: a rise of 10 °C is a rise of 18 °F, not 50 °F — the offset applies to readings, never to differences. Kelvin removes the offset problem entirely (K = C + 273.15), which is one reason science uses it. The Celsius to Fahrenheit converter and its reverse handle the formula for you; what you must supply is the judgement about whether you are converting a reading or a difference.
Worked example 1: a height, done properly
Convert 5 ft 9 in to centimetres. Step 1, collapse to one unit: 5 × 12 + 9 = 69 inches. Step 2, one multiplication by the exact factor: 69 × 2.54 = 175.26 cm. Step 3, sanity check: an average adult is 160–185 cm, so 175.26 is plausible; if you had divided instead of multiplying you would have got 27.2, which no adult human has ever been. That plausibility check — knowing roughly what the answer should look like before you compute it — catches the inverted-factor error, which is the most common conversion error of all.
Worked example 2: fuel economy, the reciprocal trap
US cars quote miles per gallon; most of the world uses litres per 100 km. These are reciprocals, not multiples: L/100km = 235.215 ÷ mpg. A car rated at 30 mpg uses 235.215 ÷ 30 = 7.84 L/100km. Note the direction flips: a bigger mpg number is better, a bigger L/100km number is worse. Any conversion where "more" becomes "less" is a reciprocal, and multiplying by a factor instead will give you a confidently wrong answer. (If you chained it the long way — miles to km, gallons to litres, then normalised to 100 km — the cancellation method from the first section gets you to the same 235.215 constant, which is how you know it is right.)
Worked example 3: the recipe that crosses the Atlantic
An American recipe calls for 2 cups of flour. Cups measure volume; your kitchen scale measures mass; the bridge between them is density, and it is different for every ingredient. For plain flour, 1 US cup ≈ 120–125 g, so 2 cups ≈ 250 g. The same 2 cups of granulated sugar is about 400 g, and of cocoa powder about 170 g. This is why "1 cup = 240 ml" — true as a volume statement — cannot finish the job: converting volume to mass without the ingredient is a category error, the domestic version of the orbiter's mistake. When a conversion needs a property of the material (density, in this case), any tool or table that gives you a single universal answer is guessing. Use an ingredient-specific figure, weigh once, and write the gram amount on your copy of the recipe.
When it went wrong: two expensive lessons
Mars Climate Orbiter, 1999
NASA's $125 million orbiter burned up in the Martian atmosphere because one contractor's software reported thruster impulse in pound-force seconds while the navigation team expected newton-seconds — a factor of 4.45. Every individual calculation was done "correctly"; the units attached to the numbers were wrong, and nothing in the pipeline checked them. The fix that came out of the investigation was not better arithmetic. It was labelling: units travel with numbers, always, and interfaces state them explicitly.
The Gimli Glider, 1983
An Air Canada 767 ran out of fuel at 41,000 feet because the fuel calculation was done in pounds while the aircraft's gauges and the refuelling figures were in kilograms — a factor of 2.2, leaving the plane with less than half the fuel the crew believed it had. The captain, an experienced glider pilot, dead-sticked it onto a decommissioned airfield in Gimli, Manitoba, with no fatalities. The lesson repeats the orbiter's: the arithmetic was fine, the unit assumption was the failure, and a single "does that number look right?" cross-check against a known quantity (the plane's known fuel capacity in the other unit) would have caught it.
Rounding, precision and significant figures
A conversion cannot be more precise than its input. "About 5 feet" converted to 152.4 cm is false precision — the honest answer is "about 152 cm". Rules of thumb: keep the same number of significant figures as the input had; never report a converted value with more decimals than the original measurement justified; and when a conversion factor is a definition (inches, feet, pounds to metric) it is exact and does not limit you — the measurement does. In engineering and medicine, also record the factor you used. "Converted with 1 lb = 0.4536 kg" in a note has settled many later arguments.
Common mistakes checklist
- Inverted factor. Multiplying where you should divide. Defence: the plausibility check — know the rough answer first.
- US vs imperial. Gallons, fluid ounces, tons and pints differ across the Atlantic. Check which one your source means.
- Temperature offset applied to differences. A 10 °C rise is 18 °F, not 50 °F.
- Rounding at every step. Carry guard digits; round once, at the end.
- Area and volume factors. 1 m² is not 100 cm² of area in the naive sense — it is 10,000 cm², because the factor squares (100²). Volume cubes it. Square and cubic conversions are where factor errors hide.
- Unit lookalikes. Ounces (mass) vs fluid ounces (volume); tons (US 2000 lb, metric 1000 kg, imperial 2240 lb); calories vs kilocalories on food labels. Same word, different quantity.
Frequently asked questions
How do I know whether to multiply or divide?
Write the conversion as a fraction that cancels: value × (target unit / source unit). If the source units cancel and the target remains, the setup is right. Then sanity-check the size: converting to a smaller unit (metres to centimetres) must make the number bigger, and vice versa.
Why is temperature conversion different?
Because the scales start at different zeros. Fahrenheit and Celsius agree on the size relationship (a degree Celsius is 1.8 Fahrenheit degrees) but 0 °C is 32 °F, so a reading needs both a scaling and a shift. Differences between two temperatures need only the scaling.
How many decimal places should my answer have?
As many significant figures as your input had, roughly. "3 miles" becomes "4.83 km", not "4.828032 km" — the extra digits claim a precision the original measurement never had.
Is a US gallon the same as an imperial gallon?
No. A US gallon is 3.785 litres; an imperial gallon is 4.546 litres — about 20% more. The same split affects fluid ounces, pints and quarts, which is why fuel economy and recipe conversions between US and UK sources go wrong so often.
What is the most common unit conversion error?
Using the factor in the wrong direction, closely followed by mixing up same-named units (ounces vs fluid ounces, US vs imperial gallons). Both are caught by the same habit: estimate the answer's rough size before you calculate, and check the result against it.